We have built a visible multi-spectral imager (MSI) for the 1.6-m Pirka telescope of the Hokkaido University
in Hokkaido, Japan. The instrument is equipped with two liquid crystal tunable filters and a 512 × 512 pixel
EMCCD camera. One of the major purposes of this instrument is to obtain multi-spectral images (series of
narrow-band images at many different wavelengths) of the solar planets rapidly. These tunable filters are a
Lyot filter with liquid crystal variable retarders and thus can tune the transmitting wavelength rapidly without
moving parts. Their spectral ranges are 400–720 nm and 650–1100 nm and the bandwidth is typically 10 nm on
both filters. The EMCCD camera can obtain images at a frame rate of about 32 Hz, which also enables us to
improve the spatial resolution with the shift-and-add or the Lucky imaging techniques. The field of view is 3.3
× 3.3 arcmin with a pixel scale of 0.39 arcsec pixel<sup>−1</sup>. The instrument also has UBV RI-band broad-band filters
and several narrow-band filters. MSI is mounted at the f/12 Cassegrain focus of the telescope. It had the first
light on February 2011, and then have been used for several astronomical and planetary science programs as a
major facility instrument at this telescope. We describe the design, construction, integration, and performance
of this multi-spectral imager.
This paper presents a method for automated detection of liver cancer regions based on transition of density at each point obtained from multi-phase X-ray CT images. For describing transition of density, two kinds of feature vectors named Density Transition (DT) and Density Change Transition (DCT) are introduced. DCT is used for extraction of cancer candidates and DT is used for suppression of false candidates. In the experiments using 14 real abdominal CT images with cancer, it was shown that the detection rate was 100% and the number of false-positives was 0.71 regions per case.